A Virtual Machine Placement Algorithm To Reduce Energy Consumption In Data Center

With the rapid development of cloud computing,in order to meet the demand of the application service in resources and quality,as the underlying physical support,the size of data center is growing.Data centers have consumed a large amount of power,increased the operating costs of enterprises and released a large amount of carbon dioxide causing great damage on the environment.So energy saving research in data center is a hot issue in the world.Based on the situation,first of all,the paper introduces virtualization technology,one key technology of the data center,which can make software and hardware of the data center be virtualized.Then,the paper introduces the problem of virtual machine placement,and the different virtual machine placement schemes have a great influence on the energy consumption of data centers.At present,the research on the energy efficiency in data centers is mainly from the two aspects,reducing energy consumption of the physical machine and reducing the energy consumption of the network.Considering the whole data center,the paper chooses to optimize the energy consumption of the physical machine and the network equipment cooperatively to reduce the energy consumption of the data center.Aiming at the virtual machine placement problem in the data center,first of all,the paper builds the model of the physical machine and switch in the data center.And based on virtualization technology,the paper builds the optimization model,the objective is the total energy consumption of the physical machine and switch,the constraints are the CPU capacity of physical machine and the link bandwidth capacity.Then the problem is abstracted as the bin-packing problem and the paper improved best fit descending algorithm to solve the virtual machine placement problem.Combined with the communication characteristics among tasks,the paper designs the corresponding algorithm NSBDF.In the end,the paper adopts the BBDF and the NSBDF algorithm is simulated,the effectiveness of the algorithm is verified.